Water quality index, in common with many other indices systems, relates to a group of water quality parameters to a common scale and combines them into single number in accordance with a chosen method or model of computation. WQI is desired to provide assessment of water quality trends for management purposes even though it is not meant especially as an absolute measure of the degree of pollution or the actual water quality (Anonymous, 1997b).
Water quality indices were first seriously proposed and demonstrated beginning in the 1970s but were not widely utilized or accepted by agencies that monitor water quality (Cude, 2003). According to Nives (1999) WQI is a mathematical instrument used to transform large quantities of water quality data into a single number which represents the water quality level while eliminating the subjective assessments of water quality and biases of individual water quality experts. The common denominator for all water quality indices is the basic principle that a quality index must synthesize data such as analytical results by means of a simple quality vector. This method makes the information more easily and rapidly interpretable than a list of numerical values. Consequently, a water quality index is a communication tool for transmitting information. The users of this information can range anywhere from being closely associated to being distantly connected to the resource (for instance, general public, users, scientists, managers, lawmakers, engineers etc.).
Traditional reports on water quality typically consist of complex variable-by variable and water body-by-water body statistical summaries. This type of information is of value to water quality experts, but may not be meaningful to people who want to know about the state of their local water bodies. Political decision-makers, non-technical water managers and the general public usually have neither the time nor the training to study and understand a traditional, technical review of water quality data. They require concise information about those water bodies. The index also allows water quality data to be compiled and reported in a consistent manner.
A number of indices have been developed to summarize water quality data in
an easily expressible and easily understood format (Couillard and Lefebvre,
1985). Horton (1965) proposed the first Water Quality Index (WQI), a great deal
of consideration has been given to the development of index methods. The basic
differences among these indices are the way their sub-indices were developed.
Walski and Parker (1974) used an exponential function to represents the sub
indices of various quality variables. Landwehr (1979) suggested the Pearson
type 3-distribution function to represents the sub indices of all the quality
variables. Bhargava (1987) modified the exponential formula; Dinius (1987) used
power function for the majority of sub indices. Nives (1999), Swamee and Tyagi
(2000)) proposed aggregate index for water quality description. In addition
Harrison et al. (2000), Faisal et al. (2003), Ahmed et al.
(2004) and Shiow-Mey et al. (2004), each have recently modified a water
Some of the sub indices have since been incorporated into water quality indices used by agencies such as the National Sanitation Foundation (NSF) (Ahmad et al., 2004). The most important WQIs belonging to environmental departments or agencies are the National Sanitation Foundation (NSF), British Colombia Water Act Quality Index, Oregon Water quality Index, Stream Watch (Southern Indiana), Malaysian Water Quality Index, France Water Quality Index, French Creek quality index, Florida Stream water Quality index, British Colombia Water Quality Index, Canadian Water Quality index, Taiwan Water Quality index and Washington State Water Quality index.
Comparison among several WQI systems currently in use showed that none of them describes quality of water from mining effluent because most of sub indices in current WQIs are not relevant to indicating changes in water quality brought about by mining activities.
It is important to remember that, prior to this study no significant water quality information (using mining index) was available for mining or tin by product activities. This survey provides the first WQI applicable to describe the quality of water as a consequence of mining and related activities.
MATERIALS AND METHODS
Thirty three water samples were taken from seven amang (tin tailing) processing plants located in the state of Selangor and Perak, Malaysia. The amang plants comprise of three employing the open and close (natural and man made) management systems. In open water management system the water used in the processing of tin tailing is drawn from a river and the effluent from the plant is discharged directly into the same river. In the close water management system the water is recycled. All water samples were collected and analyzed in accordance with procedures outlined (Anonymous, 1997a).
Elemental analyses were carried out using ICP-MS and ICP-OES. Radionuclides activity concentration were determined using multi channel analysis integrated to a Hyper Pure Ge-Li detector. Expert input via questionnaire survey was employed to facilitate selection of parameters to be used in WQI. In this respect, a survey form was prepared and was sent to 95 scientists who are experts in water and wastewater treatment around the world. The survey asked respondent the expert opinion to help formulate possible new sub-indices for measuring WQI specific measuring water quality as a consequent to mining and mineral processing. Respondents are requested to select and ranked water quality parameters listed that they consider critical and should be included in the equation to calculate this WQI.
Proposal of a new WQI: The main objective of this study was to propose a new water quality index applicable to mining effluent. Generally, there are five basic steps involved in the development of most water quality indices. These include:
||Selecting the set of water quality variables (parameters)
||Aggregation or designing framework of formula.
||Weighting the indicators based on their relative importance
to overall water quality.
||Developing rating curves for comparing indicators on a common
||Computing the overall water quality index
RESULTS AND DISCUSSION
Selecting the set of water quality variables: Variables of concern to water quality should be selected from several commonly recognized impairment categories including health aspect, physical and chemical characteristics. In each WQI, these selected parameters could be varying. In this study, a wide measurement of parameters in whole water samples such as physical, chemical and environmental parameters (including heavy metals and radionuclides) were carried out. Table 1 showed that the important parameters in amang processing effluent water samples (based on chemical reactions and physical interactions in amang processing) were pH, PO43¯, NO3¯, electrical conductivity, total solid, solid suspended, total dissolved solid, dissolved oxygen, SiO2, radionuclides (U-238 and Th-232), some elements (Na, Mg, K, Ca, Al) and some heavy metals (Mn, Fe, Ni, Cu, Zn, As, Se, Cd, Pb).
Table 2 ranked the 10 most preferred parameters perceived by respondents to be important in calculating water quality index. Table showed that the most important environmental parameter or indicator is pH (38 points).
||Comparisons between average water quality parameters of sampled
water and INWQS/Canadian and USEPA standards
|*: Interim National Water Quality index (INWQS)-Malaysia,
**: Canadian standard; ***: United State Environmental Protection Agency
||Selected parameters and points scored based on respondents
The other nine parameters were NO3¯ (28 points),
radionuclides 238U and 232Th (26 points), DO (24 points),
PO43¯ (23 points), heavy metals (23 points),
Electrical Conductivity (18 points), SO42¯ (17
points), Suspended Solid (16 points) and Total Dissolved Solid (13 points).
A WQI should be made up of sub indices of parameters that could explain the
quality of water comprehensively. Redundancy should be avoided. The survey showed
that both Total Dissolved Solids (TDS) and electrical conductivity were selected
in the top 10 preferred parameters. Since there is apparent and direct relationship
between total dissolved solid and electrical conductivity which was ranked 6th
and 10th, respectively it was decided that electrical conductivity was preferred
over TDS to be included in the WQI calculation.
Aggregation or designing skeleton of formula: The new WQI proposed is in accordance to Nives (1999). The WQI was calculated by summing up individual quality rating (qi) and weighting these parameters in total quality evaluation (wi) as shown in Eq. 1.
||Water quality score of parameter i.
||Weighting factor of parameter.
|i and n
||No. of parameters.
Based on nine selected parameters the new WQI proposed is as shown in Eq. 2.
Where, A, B, C, D, E, F, G, H and I are weighting factors for the respective sub indices. These weighting factors shall indicate the importance of the related sub indices toward the overall WQI.
The Sub Index (SI) of each parameter is obtained from related rating curves or equations. The main advantage of a rating carve is that it rapidly transformed the concentration or quotient of a parameter into a quality score i.e., Sub-Index (SI). Axes in rating curve consist of X and Y axes, representing concentration or quotient of the parameter tested and corresponding SI, respectively.
Sub index for radionuclides: For radionuclides, r(average) is calculated using Eq. 3:
Where, Σ[ri] is the summation of ri and ri is the activity concentration quotient of radionuclide i, while n is the number of radionuclides considered in the calculation.
||Concentration of radionuclide i in sample
||Maximum Permissible Activity Concentration of radionuclide i.
The Maximum Permissible Activity Concentrations of radionuclides are provided in the National Water Quality Standard.
Due to the toxicity of radionuclides, a limit for ri is introduced. SI for radionuclides becomes zero if any one radionuclide quotient,
In such a case, the SI for radionuclides equal zero. The SI for corresponding
r(average) is obtained from Fig. 5.
Sub index for heavy metals: Similar mathematical approach to that in deriving the SI of radionuclide is used for heavy metals. Again for heavy metals h(average) can be calculated according to formula Eq. 5:
||The summation of hi.
||The concentration quotient of element i.
||The number of elements considered.
||The concentration of element i (heavy metal) in sample and
||Maximum permissible concentration of element i (According to Water
The national authority shall determine selection for the number of heavy metals to be considered in the calculation of SI. However, as a result of toxicity of heavy metals, there is a limitation where the SI (heavy metals) shall be forced to become zero. It is zero, if and only if at least one of the quotients of concentrations (hi) is equal or more than 1. In other word SI for heavy metals, equals to zero when the concentration of at least one heavy metals in the water sample is larger than its maximum permissible concentration, i.e.,
The SI for corresponding h(average) is obtained from Fig. 6.
Weighting factors for SI: Weighting factors indicate the importance of each test parameters towards the overall water quality. In the case of the new WQI, the weighting factors for pH, nitrates, radionuclides, dissolved oxygen, phosphates, heavy metals, electrical conductivity, sulphates and suspended solids are A, B, C, D, E, F, G, H and I, respectively (Eq. 2). Weighting factors for A-I were calculated based on the frequency of respondents that selected each test parameters (Table 2) and then weighting factors were normalized. The calculated frequency (in fraction) for each test parameters is integrated into equation 2 to produce a new WQI (Eq. 7).
Developing rating curves for SIS: This step involves the transformation of all test parameters to an equal and dimensionless scale. This is generally accomplished using rating curve, where each test parameter concentrations or quotients are mapped against a dimensionless measure such as relative water quality value or SI.
Rating curves has to be developed for all test parameters. These rating curves
have been developed using existing rating curves, or from secondary data and
in consultations with experts in the field of water quality. The rating curves
for pH, dissolved oxygen, nitrate and phosphate were taken directly from the
curves developed by National Sanitation Foundation (NSF). The rating curve for
radionuclides, heavy metals and also for sulfate and electrical conductivity
were developed and designed through reviewing of literatures on the relationships
between water quality and known concentrations of elements. In addition, Maximum
Permissible Value (MPV) and Maximum Permissible Standards (MPS) information
belonging to United State Environmental Protection Agency (USEPA, 2002), classification
of water prescribed by the Department of Environmental of Malaysia (Anonymous,
1998) and Canadian quality standards (Faisal, 2003) were considered in the development
of the rating curves. Table 3 shows list of drinking water
contaminants according to USEPA, DOE Malaysia and Canadian quality standards.
Rating curves developed for use with the proposed WQI are as shown in Fig.
|| List of water quality standards
*: MCLG: Maximum contamination level goal- The level of the
contaminant in drinking water below which there is no known or expected
risk to health. **: MCL: Maximum contamination level- The highest level
of a contaminant that is allowed in drinking water. ***: According to US
Environmental Protection Agency there is no value (amount) in National Primary
Drinking water Regulation and these values are belonging to National Secondary
Drinking Water Regulations
As mentioned earlier, case of this study is related to effluent from amang (tin tailing) processing activity. In this case seven heavy metals are recommended and considered in calculating sub index for heavy metals. These heavy metals are Cd, Pb, As, Hg, Ni, Mn and Fe. Selection of these heavy metals is related to high concentration of these elements in amang effluent and their high toxicity.
Similarly uranium-238 and thorium-232 were considered as the most important radionuclides because of their high concentrations in amang and its effluent. In addition amang processing has been shown to enhance these naturally occurring radionuclides (Ismail et al., 2003).
The Canadian water quality standard was used for calculating ratio (ri).
Canadian water quality standards was used because their MPLs were very close
to standard values belonging to USEPA (2002) and are lower than the values used
by the DOE, Malaysia. It should also be mentioned that based on the Canadian
standard, the high quality of water or the best class of water belongs to those
use for drinking water followed by aquatic uses and finally agriculture uses.
In this study agriculture values (lowest class) belonging to Canadian water
quality standards were used as maximum permissible value for heavy metals and
calculating concentration ratio (Table 4).
SI for WQI parameters may be determined using related rating curves or using related fit Equation(s).
SI for pH: In accordance with Fig. 1 and reference to WQI of the DOE-Malaysia, The SI for pH is estimated as follows:
||Maximum permissible levels (MPLs) used in this study for the
development of sub index and WQI
If 2 < pH < 12 then above equations may be substituted with a single best fit equation of the rating curves (Fig. 1) as follows:
Where, X is pH.
SI for NO3¯: SI for NO3¯ may be determined using rating curve shown in Fig. 2 or using a best fit equation of the rating curve as shown in Eq. 15:
Where, X is concentration of NO3¯ (mg L1)
SI for DO: SI for Dissolve Oxygen (DO) may be calculated using the best fit equation for the estimation of sub-index value for dissolved oxygen as proposed by DOE-Malaysia:
Where, X is dissolved oxygen. (% saturation)
SI for DO may also be calculated using the best fit equation for the rating curve shown in Fig. 3 (Eq. 21).
|| pH rating curve
|| Nitrate rating curve
Where, X is dissolved oxygen (% saturation)
SI for PO43¯: The best-fit equations (Eq.
22 and 23) for the estimation of sub index value for
phosphate are derived from the curve shown in Fig. 4.
|| DO rating curve
|| Total phosphate rating curve
Where, X is concentrations of PO43¯ (mg L1)
Direct estimation of SI for may also be made directly from the rating curve shown in Fig. 4.
SI radionuclides: SI for radionuclides is estimated using Fig.
5, or estimated using the best-fit Eq. 24 and 25.
SI for radionuclides is estimated based on activity concentration quotients
(ri). The use of activity concentration quotients (ri)
for radionuclides allow the use of 1 rating curve for all radionuclides. This
is made possible because the determination of SI is based on relative concentrations
of measured values and those of permissible concentrations as prescribed in
the national water quality standards.
|| Radionuclides rating curve
Examples of some of these maximum permissible values are shown in Table
3. Equation 24 is for estimating SIradionuclides when
r(average) is less or equal to one, equation 25 is for estimating SI radionuclides
when (ri) is greater than one.
A linear inverse relationship is proposed for SI and ri. Such relationship is preferred over up or down bending curves, because an up bending curve may not be economical to implement, while down bending curve allows for unnecessary underestimation of risk (Fig. 5).
Where, X is r(average).
SI for heavy metals: Similar to calculating SI for radionuclides, SI for heavy metals is estimated based on concentration quotients hi. Similar to the rationale used in determining SI for radionuclides, the use of concentration quotients for heavy metals allow the use of one rating curve for all heavy metals. Equation 26 is for estimating SI heavy metals when h(average) is less than one. Equation 27 is used for estimating SI for heavy metals when h(average) is equal or greater than one (Fig. 6).
Similar argument in using an inverse linear relationship between SI and hi, over up and down bending curve for radionuclides applies for heavy metals (Fig. 6). SI for heavy metal may also be estimated using a best-fit equation derived from Fig. 6 (Eq. 26).
|| Electrical conductivity rating curve
Where, X is h(average).
SI for electrical conductivity: This rating curve was developed according to standard values derived from both the Canadian and Malaysian water quality standards since their standards complement each other. According to Canadian water quality index maximum permissible level of electrical conductivity for drinking water is 700 μs cm1 and according to DOE Malaysia, maximum permissible level of electrical conductivity in class IV (lowest quality) is 6000 μs cm1. The interpretation of Fig. 7 is that, when the electrical conductivity rises to 700 μs cm1 (max permissible level for drinking water), the SI conductivity comes down to 50 and when the electrical conductivity rises to 6000 μs cm1 (max permissible in class IV), the SI conductivity comes down to zero.
|| Sulfate rating curve
The best-fit equations for estimating SI for electrical conductivity are as
shown in Eq. 28 and 29.
Where, X is electrical conductivity (μs cm1).
SI for sulfate: This rating curve was developed based on values drawn from the Canadian and Malaysian water quality standards. Best-fit equation for this rating curve is as shown in Eq. 31. When concentration of sulfate is zero, SI sulfate becomes 100 and when concentration of sulfate rises to 500 mg L1 (maximum permissible value in drinking water according to Canadian standard) SI Sulfate becomes 50 and when concentration of sulfate rises to 1000 mg L1 (maximum permissible value in agriculture uses, according to Canadian standard), SI sulfate becomes zero. An inverse linear relationship between SI and sulfate concentration is proposed to avoid under and over estimation in determining SI for sulfate (Fig. 8).
In accordance with Fig. 8, the best fit Equation for the estimation of sub index value for sulfate is:
Where, X is concentration of sulfate (mg L1).
SI for suspended solid: Rating curve for suspended solid was developed
based on the Malaysian Interim National Water Quality Standards. The Standards
proposed two levels of suspended solid according to the classification of water,
i.e., 20 and 150 mg L1 for Class I and III, respectively. As such
three equations are proposed in calculating SI for suspended solids. Equation
33 is for estimating SI for suspended solid concentration <20 mg L1,
Eq. 34 is for estimating SI for suspended solid concentrations
between for 20 and 150 mg L1. SI equals zero if the concentration
of suspended solids is greater than 150 mg L1.
According to Fig. 9, the best-fit equations for the estimation of sub index value for suspended solid are:
Where, X is concentration of suspended solid (mg L1).
Application of new WQI to existing data collected: The new WQI was tested using data collected from amang effluent. Result from one station (Table 5) was used to illustrate this. A score of 31.81 classifies the water as very polluted (Table 6).
In this study Malaysian water quality rating scale was employed for classification of water. There are three rating scales (clean, slightly polluted and very polluted) and five water classes (Table 6).
It is important to remember that, prior to this study no specific water quality
index to assess the quality of water as a consequence of mining activities.
Furthermore, no attempt to study the comparative water quality of large numbers
of water samples in amang industries has previously been made. Ahmed et al.
(2004) proposed a water quality index. However this water quality index cannot
be used to indicate contamination from trace metals, organic contaminants, or
other toxic substances. Shiow et al. (2004) also has proposed a new water
quality index but this index mainly focuses on the organic pollution caused
by the municipal and agriculture activities. Amang water contains heavy metals
and radionuclides which is not only chemically toxic but also posed a radiological
As mentioned earlier, each WQIs has strength and weakness. For example, the
current Malaysian Water Quality Index (WQI) is acceptably defined by integrating
six sub indices, i.e., dissolved oxygen, pH, BOD, COD, ammoniacal nitrogen and
total suspended solid. Unfortunately such definitions of water quality have
been loosely used to describe all types of water independent of whether it is
domestic or mining effluent. The Malaysian WQI is not appropriate to describes
water quality as a consequence of mining, because the sub indices used in manipulating
the overall WQI are only applicable to domestic or mostly related to organic
wastewater but is not for mining effluent. Parameters such as COD, BOD and Ammoniacal
Nitrogen (AN) are neither momentous nor relevant in describing changes in the
quality of water brought about by mining effluent.
|| Suspended solid rating curve
|| Computing the overall water quality index based on the new
|*: WQI =0.18 (SI pH) +0.13 (SI NO3¯) +0.12
(SI radionuclides) +0.11 (SI DO) + 0.11 (SI PO43¯)
+0.11 (SI heavy metals) + 0.08 (SI electrical conductivity) + 0.08 (SI SO42¯)
+ 0.08 (SI solid suspended); **: Refer to related tables
|| Water quality index classifications
|Source: Anonymous (1997)
The Malaysian water quality index, uses higher constant values of SI not relevant
to mining effluent (e.g., BOD, COD and AN) that masked the contribution of changes
in SISS and SIpH towards the overall WQI. Current WQIs, also do not take into
consideration contribution from inorganic materials and naturally occurring
radionuclides present in mining. A review of more than 20 WQIs revealed that
not one of these WQIs formulas can illustrate accurate water quality index related
to tin mining and amang processing activities.
Some WQIs, for example the Oregon Water Quality (OWQI) comes with limitations in using its index. The OWQI aids in the assessment of water quality for general recreational uses (i.e., fishing and swimming). The OWQI cannot determine the quality of water for specific uses, nor can it be used to provide definitive information about water quality without considering all appropriate chemical, biological and physical data. The OWQI was designed for Oregons streams and its applications to other geographic regions or water body types should be approached with caution (Cude, 2003).
The introduction of a new water quality index especially for mining provides a fairer assessment of water quality brought about by mining circumstances (i.e., mining and by product processing activities).
Water quality classes are useful for summarizing information in order to obtain regional and national perspective. Data collected from current study have shown that the current WQIs used to gauge water quality are not appropriate for use to determine the quality of water related to amang processing. This is because the sub indices used in calculating the overall WQI is more applicable for domestic effluent but not for mining effluent. For instance, in current WQIs parameters such as DO, COD, BOD and Ammoniacal Nitrogen (AN) were not significantly affected in amang processing activities. Consequently even though, the water quality was affected in terms of its pH and total suspended solid, the overall quality of water as measured using the current WQI was not affected.
In this study a wide parameters or environmental tests have been tested including physical, chemical, organic and inorganic materials, heavy metals and radionuclides in the process of proposing an appropriate water quality index in mining and especially in amang activities. Finally an amended WQI was introduced integrating parameters relevant to such activities. Nine commonly water quality indicators provide adequate coverage of traditional impairment categories. Water bodies adjacent to amang industries can be effectively summarized using proposed new WQI.
The new proposed WQI has some advantages over other WQIs. The most important advantage of this water quality index is utilizing of relevant and sufficient variables for identifying quality of water from amang processing plants. This index contains nine variables similar in number to those used by NSF quality index. This index gives results very similar to those calculated using NSF. In addition, the newly proposed WQI took greater emphasis on the contributions of heavy metals and radionuclides.